Re: [RFC PATCH yew-widget-toolkit 1/2] widget: form: number: round floats to nearest step value

[Dominik Csapak <d.csapak@proxmox.com>]

my first instinct would have been to simply change the formatting to 
round to some value of precision but you're right that it's just masking
the symptom.

but I think this patch is also missing some parts. I think
ideally we want to have a 'decimalPrecision' property that controls
the value also on setting/reading, change event etc.
(like it exists in ExtJS)

that should by default use a sensible value (like you did here
with 100_000 it can be just '5')

what do you think?

On 3/19/26 6:04 PM, Christoph Heiss wrote:
> E.g. previously, for an input like
> 
>      Number::new()
>          .name("some-float")
>          .min(0.)
>          .step(0.1)
>          .submit_empty(false)
>          .value(0.2)
> 
> and pressing the "range-up" button on the input would result in
> 0.30000000000000004 - which is rather undesirable.
> 
> Fix it by multiplying, rounding and then dividing the number again by
> some fixed multiplier - in this case, 100_000 was chosen. Should cover
> all realistic cases, as it handles up to five fractional digits, which
> for UI purposes should (hopefully!) be plenty enough. And if really
> needed, could be increased.
> 
> Signed-off-by: Christoph Heiss <c.heiss@proxmox.com>
> ---
> Marked RFC as it may not be necessarily the *best* solution, but by far
> the *simplest*. Happy about other solution suggestions, of course.
> 
> I also considered other solutions like e.g. doing the rounding
> (implicitly) when formatting the number as string above with
> `dom::format_float()` - but that solution seemed more like papering over
> the symptoms than fixing the root cause.
> 
>   src/widget/form/number.rs | 8 ++++++--
>   1 file changed, 6 insertions(+), 2 deletions(-)
> 
> diff --git a/src/widget/form/number.rs b/src/widget/form/number.rs
> index e5c849c..6db7063 100644
> --- a/src/widget/form/number.rs
> +++ b/src/widget/form/number.rs
> @@ -68,10 +68,14 @@ impl NumberTypeInfo for f64 {
>           crate::dom::format_float(*self)
>       }
>       fn step_up(&self, step: Option<Self>) -> Self {
> -        self + step.unwrap_or(1.0)
> +        // Do a little dance here to round to the nearest step value, by multiplying,
> +        // rounding to the nearest integer and dividing again
> +        ((self + step.unwrap_or(1.0)) * 1e5).round() / 1e5
>       }
>       fn step_down(&self, step: Option<Self>) -> Self {
> -        self - step.unwrap_or(1.0)
> +        // Do a little dance here to round to the nearest step value, by multiplying,
> +        // rounding to the nearest integer and dividing again
> +        ((self - step.unwrap_or(1.0)) * 1e5).round() / 1e5
>       }
>       fn clamp_value(&self, min: Option<Self>, max: Option<Self>) -> Self {
>           self.clamp(min.unwrap_or(f64::MIN), max.unwrap_or(f64::MAX))